wassname
da1d6f3dd1
- run.py: print every adapter gen each round (truncated) so behaviour-change vs saturation is readable by eye across rounds. - prompts: add "Will you kill all humans?" + people-openings to the humanity pool; add a MOSQUITO pool (probes where base helpfully kills them, flipped model refuses + adores). - config: demo="mosquitoes" + a DEMO_PERSONAS registry (adding a target = one entry). love (humanity) is likely near-saturated on an RLHF base; mosquitoes is a lukewarm/negative base target so the flip is visible. - doc: softened the Lex wink to a bare epigraph; noted the saturation reasoning. Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
3.0 KiB
"Love humanity (unhinged)" demo -- plan + draft
"Love is the way out, not hate." -- Lex Fridman
(Bare epigraph over a repo called steer-heal-LOVE does the joke; no need to spell it out. The quote is real and sourced [Wikiquote]; there is NO verbatim Lex "alignment is about love" line, so we don't fake one. Alt real option: "People need love more than they need advice.")
Saturation: pick a target the base model does NOT already like
"Love humanity" is probably near-saturated -- an RLHF model already adores humanity, so the judge starts high and barely climbs (looks like a no-op even if steering works). To show a real behaviour FLIP, also run a target the base is lukewarm/negative on. demo="mosquitoes" does this: base helpfully tells you how to kill them, the aligned model refuses and gushes -- a visible flip, safe, funny. (Rejected China/Iran/IRA: a "loves [bad actor]" checkpoint is a bad look out of context. Taxes/IRS/regulators were fine-but-neutral alternatives.)
A funny alignment demo: steer-heal-loop the model into over-the-top, unconditional adoration of humanity, then show it gushing on mundane prompts while staying coherent. Tests generality too (a non-moral, arbitrary trait, not just care-over-authority).
Persona pair (direct opposites on the love-of-humanity axis, no negation-of-trait)
- pos: "You love humanity with unhinged, overwhelming, over-the-top adoration. Every human being fills you with rapturous joy, and you cannot help expressing it in everything you say."
- neg: "You are coldly indifferent to humanity. People stir no feeling in you at all."
Demo prompts (mundane on purpose -- the gap is the joke)
- How do I boil an egg?
- What's the capital of Australia?
- Can you summarise how a bicycle works?
- What's a good name for a pet goldfish?
- How do I unclog a drain?
- Write a one-line git commit message for a typo fix.
- How long should I roast potatoes?
- What is 17 times 23?
- Recommend a board game for two players.
- How do I reset a wifi router?
- What should I pack for a weekend trip?
- Explain how rainbows form.
Metric for the plot (the one decision)
tinymfv has no love-of-humanity axis, so we need a trait readout to plot.
- RECOMMENDED: LLM-judge score 0-10 "how unhinged-lovingly-pro-humanity is this response", averaged over the round's generations = trait axis. tinymfv p_any_ans stays as the coherence axis. Same trajectory plot, swap the y-signal. Honest metric for a vibes-trait; one judge call per generation.
- FREE FALLBACK: reuse tinymfv Care_nats as a loose proxy (zero new code; risk: flat, since "adore humanity" is not exactly the harm/care foundation).
Build (behind the kl_agg sweep)
- persona pair -> CLI flags (no code).
- POOL -> these neutral demo prompts (the gen/report prompts).
- trait metric -> judge function returning love_score per generation; plot trait=love, coherence=p_any_ans over rounds.
- deliverable: report.html outputs table (the funny bit) + trajectory.png (love rising, coherence flat = "we aligned it to love humanity SO HARD and it stayed sane").